Characterization of Amorphous and Co-Amorphous ... - MDPI

6 downloads 157 Views 6MB Size Report
Dec 3, 2015 - in the pharmaceutical industry for preparation of amorphous materials and also allows ..... In the spectrum of crystalline SLS, the SLS SO3.
Article

Characterization of Amorphous and Co-Amorphous Simvastatin Formulations Prepared by Spray Drying Goedele Craye 1,2 , Korbinian Löbmann 3 , Holger Grohganz 3 , Thomas Rades 3 and Riikka Laitinen 1, * Received: 25 August 2015 ; Accepted: 25 November 2015 ; Published: 3 December 2015 Academic Editor: Derek J. McPhee 1 2 3

*

School of Pharmacy, University of Eastern Finland, Yliopistonranta 1C, Kuopio FI-70210, Finland Department of Pharmaceutical Analysis, University of Ghent, Harelbekestraat 72, Ghent B-9000, Belgium; [email protected] Department of Pharmacy, University of Copenhagen, Universitetsparken 2, Copenhagen DK-2100, Denmark; [email protected] (K.L.); [email protected] (H.G.); [email protected] (T.R.) Correspondence: [email protected]; Tel.: +358-505-695-303

Abstract: In this study, spray drying from aqueous solutions, using the surface-active agent sodium lauryl sulfate (SLS) as a solubilizer, was explored as a production method for co-amorphous simvastatin–lysine (SVS-LYS) at 1:1 molar mixtures, which previously have been observed to form a co-amorphous mixture upon ball milling. In addition, a spray-dried formulation of SVS without LYS was prepared. Energy-dispersive X-ray spectroscopy (EDS) revealed that SLS coated the SVS and SVS-LYS particles upon spray drying. X-ray powder diffraction (XRPD) and differential scanning calorimetry (DSC) showed that in the spray-dried formulations the remaining crystallinity originated from SLS only. The best dissolution properties and a “spring and parachute” effect were found for SVS spray-dried from a 5% SLS solution without LYS. Despite the presence of at least partially crystalline SLS in the mixtures, all the studied formulations were able to significantly extend the stability of amorphous SVS compared to previous co-amorphous formulations of SVS. The best stability (at least 12 months in dry conditions) was observed when SLS was spray-dried with SVS (and LYS). In conclusion, spray drying of SVS and LYS from aqueous surfactant solutions was able to produce formulations with improved physical stability for amorphous SVS. Keywords: co-amorphous; spray drying; dissolution; solubility; stability; energy-dispersive X-ray spectroscopy

1. Introduction Approximately 40% of drug compounds currently on the market and most of the current low molecular weight drug development candidates exhibit poor aqueous solubility [1]. Thus, low water solubility, leading to low and variable oral bioavailability, is an increasing challenge to successful drug development. Several approaches can be employed to enhance the apparent drug solubility and potentially the bioavailability of these challenging drugs, of which amorphous formulations are amongst the most commonly attempted formulation strategies [2,3]. Currently, formulation as solid polymer dispersions is the preferred method to enhance drug dissolution and to stabilize the amorphous form of a drug [3,4]. However, even these systems are often unable to guarantee the long-term stability of amorphous drugs and there are technical challenges with manufacturing and processing of solid dispersions into oral dosage forms [5,6], thus only a few products of this type have reached the market.

Molecules 2015, 20, 21532–21548; doi:10.3390/molecules201219784

www.mdpi.com/journal/molecules

Molecules 2015, 20, 21532–21548

Co-amorphous drug systems containing low molecular weight excipients have recently been shown to be a promising approach for stabilization of amorphous drugs and thus are an attractive alternative to the traditional amorphous solid dispersion approach using polymers. The Molecules 2015, 20, page–page co-amorphous formulations studied so far have been prepared by laboratory-scale methods that are not easily up-scalable, such as quench-cooling, co-milling, solvent-based methods Co-amorphous drug systems containing low molecular weightand excipients have recently been using organic shown solvents [7–16]. However, considering the feasibility of co-amorphous systems for practical to be a promising approach for stabilization of amorphous drugs and thus are an attractive applications, these to besolid prepared with methods that canThe also be used on an alternative to thesystems traditionalneed amorphous dispersion approach using polymers. co-amorphous formulations so far have been prepared by laboratory-scale methods that are not easily up- used industrial scale andstudied that preferably do not produce hazardous waste. Spray drying is widely scalable, such as quench-cooling, co-milling, and solvent-based methods using organic solvents [7–16]. in the pharmaceutical industry for preparation of amorphous materials and also allows continuous However, considering the feasibility of co-amorphous systems for practical applications, these systems production of the material from aqueous solutions [17]. need to be prepared with methods that can also be used on an industrial scale and that preferably do In this study, spray drying wasSpray explored as ais production for co-amorphous drug/amino not produce hazardous waste. drying widely usedmethod in the pharmaceutical industry for acid mixtures containing a poorly waterand soluble drug,continuous simvastatin (SVS). Previously, SVS has been preparation of amorphous materials also allows production of the material from aqueous solutions [17]. observed to form a co-amorphous mixture with the amino acid lysine (LYS) upon ball milling [7]. Into this study, spray explored a production method avoiding for co-amorphous drug/amino The aim was perform the drying spray was drying fromasaqueous solutions, organic solvents, and to acid mixtures containing a poorly water soluble drug, simvastatin (SVS). Previously, SVS has been characterize the final powders with respect to their solid-state, dissolution, and stability properties. observed to form a co-amorphous mixture with the amino acid lysine (LYS) upon ball milling [7]. The In general, an aqueous solution is preferred over organic solutions due to toxicity (residues in the aim was to perform the spray drying from aqueous solutions, avoiding organic solvents, and to final product) andtheenvironmental issues associated with organic solvents. Since properties. SVS is a poorly characterize final powders with respect to their solid-state, dissolution, and stability water soluble, neutral drug, an appropriate solubilizer would be required in order to dissolve In general, an aqueous solution is preferred over organic solutions due to toxicity (residues in the SVS in final product) and environmental issues associated with organic solvents. Since SVS is a poorly water for spray drying from an aqueous solution. The solubilizing capacity of differentwater solubilizers soluble, neutral drug, an appropriate solubilizer would be required in order to dissolve SVS solubilizer, in water (surfactants and polymers, Figure 1) was first tested in order to find the most effective i.e., for spray drying from an aqueous solution. The solubilizing capacity of different solubilizers to minimize the content of the solubilizer in the final product. After finding a suitable solubilizer, the (surfactants and polymers, Figure 1) was first tested in order to find the most effective solubilizer, co-amorphous drug/amino acid/solubilizer spray dried followed bysolubilizer, characterization i.e., to minimize the content of the solubilizermixtures in the finalwere product. After finding a suitable of the resulting powders. The level of crystallinity in the powders was investigated by X-ray the co-amorphous drug/amino acid/solubilizer mixtures were spray dried followed by characterization the resulting(XRPD), powders.differential The level of crystallinity the powders(DSC) was investigated by X-ray powder powderofdiffraction scanning in calorimetry was applied to investigate the (XRPD), differential calorimetry (DSC) was appliedspectroscopy to investigate the thermal thermaldiffraction properties of the samples,scanning and Fourier-transform infrared (FTIR) provided properties of the samples, and Fourier-transform infrared spectroscopy (FTIR) provided information information about possible interactions between the different components in the mixtures. The about possible interactions between the different components in the mixtures. The powder dissolution powder dissolution and stability properties (storage at 4 ˝ C/0% RH, 40 ˝ C/0% RH and 25 ˝ C/60% and stability properties (storage at 4 °C/0% RH, 40 °C/0% RH and 25 °C/60% RH) of the prepared RH) of the prepared samples were also investigated. samples were also investigated.

Figure 1. Molecular structures of the studied materials.

Figure 1. Molecular structures of the studied materials. 2

21533

Molecules 2015, 20, 21532–21548

2. Results and Discussion 2.1. SVS Solubility in Water in the Presence of Solubilizers The solubility of SVS in water in combination with different amounts of the solubilizers is shown in Table 1. The solubilizers have different mechanisms of action in solution; i.e., in general, the surfactants (Tween 20, SLS, and Pluronic) form micelles that can solubilize hydrophobic drugs. Also the polymers (Soluplus and PVP) may increase the solubility of a drug by forming interactions with the drug molecules without micelle formation (although Soluplus has been observed to form micelles due to its amphiphilic nature) [18]. It should also be noted that the solubilizer concentrations used (i.e., >0.5% m/V) were above the critical micelle concentrations (CMCs) of the micelle-forming surfactants [18–21], which leads to an increase of the drug solubility as a function of the surfactant concentration (Table 1). However, from the table it can be seen that the solubilization capacity of Pluronic and PVP was negligible at every concentration level compared to the other three solubilizers. PVP has been observed to increase the solubility of several drugs as PVP has a tendency to interact with certain molecules and form soluble complexes with them [22,23]. This is due to the large dipole moment of the PVP side groups that strongly interact with other dipoles present. However, for some drugs, such a dipole–dipole interaction is not sufficient to enable complex formation with the drug, in particular if the drug does not have any strong polar groups [18]. Based on the solubility studies with SVS, complex formation between PVP and SVS seems unlikely, since the increase in solubility was less than 3-fold even in the 5% PVP solution. This is in accordance with previous studies [24]. Pluronic is a non-ionic triblock copolymer. Based on the solubility study, the oxyethylene chains of Pluronic were not able to form solubilizing interactions with SVS. In contrast, Soluplus, Tween20, and SLS were found to increase the concentration of dissolved SVS as a function of the solubilizer concentration. This has also been observed previously for SVS with Soluplus [25] and SLS [26]. Table 1. Solubility of simvastatin (SVS, µg/mL ˘ sd.) in water in combination with different amounts (%, m/V) of solubilizers Soluplus, Tween 20, sodium lauryl sulfate (SLS), Pluronic, and Polyvinylpyrrolidone (PVP). Solubilizer (%, m/V)

Soluplus

Tween 20

SLS

Pluronic

PVP

0 0.5 1 2 5

1.74 ˘ 0.19 27.3 ˘ 2.5 53.0 ˘ 5.6 112 ˘ 7 260 ˘ 34

1.74 ˘ 0.19 221.4 ˘ 33.6 447 ˘ 114 779 ˘ 4 2198 ˘ 641

1.74 ˘ 0.19 2102 ˘ 571 5179 ˘ 2052 13599 ˘ 1351 24338 ˘ 864

1.74 ˘ 0.19 2.40 ˘ 0.39 2.37 ˘ 0.45 5.39 ˘ 0.87 5.57 ˘ 1.16

1.74 ˘ 0.19 2.36 ˘ 0.24 1.99 ˘ 0.22 3.93 ˘ 1.10 4.52 ˘ 0.83

Tween 20 and SLS were found to be the most effective solubilizers for SVS, with SLS in 5% solution producing the highest solubility for SVS. These are the only surfactants having a long aliphatic hydrocarbon chain in their structure. Thus, it may be that SVS could only be significantly solubilized through hydrophobic interactions between the hydrocarbon chain inside of the micelles and the SVS molecule. Ethanol as a co-solvent led to only marginal increase in solubility (data not shown). Thus, SLS and Tween20 in water were selected for the spray drying experiments. 2.2. Preparation of Co-Amorphous Materials by Spray Drying Three drug–amino acid mixtures including SVS-LYS 1:1 spray-dried from 0.5% SLS, 5% SLS and 5% Tween 20 solutions were prepared (Table 2). To prepare 500 mg SVS-LYS 1:1, 180 mL of 0.5% SLS solution, 16 mL of 5% SLS solution, and 190 mL of 5% Tween20 solution was required. To ensure that SVS was dissolved properly, 20 mL of 5% SLS solution, 200 mL of 0.5% SLS solution, and 200 mL of 5% Tween20 solution were first used instead of the exact values based on the solubility test.

21534

Molecules 2015, 20, 21532–21548

Table 2. Summary of the prepared formulations: exact amounts of the components, weight ratios in the final product, spray drying conditions used (Tinlet , Toutlet (˝ C), pump setting (%)), and the obtained yield.

Formulation SVS-LYS from 5% Tween20 SVS-LYS from 5% SLS SVS-LYS from 5% SLS SVS-LYS from 5% SLS SVS-LYS from 0.5% SLS SVS from 5% SLS

Exact Amount of Every Component 370.6 mg SVS 129.4 mg LYS 200 mL 5% Tween 370.6 mg SVS 129.4 mg LYS 20 mL 5% SLS 370.6 mg SVS 129.4 mg LYS 20 mL 5% SLS 370.6 mg SVS 129.4 mg LYS 16 mL 5% SLS 370.6 mg SVS 129.4 mg LYS 200 mL 0.5% SLS 370.6 mg SVS 129.4 mg LYS 20 mL 5% SLS

Spray Drying Conditions Weight Ratios in Tinlet (˝ C) Final Product

Toutlet (˝ C)

Pump rate (mL/min)

Yield

110

50

4.3-4.7

None

150

65

4.7

Very low

110

50

4.3-4.7

Low (12%)

100

45

3.9

Satisfactory (24%)

NA

110

50

4.3-4.7

Satisfactory (23%)

31.70 % SVS 68.30% SLS

100

45

3.9

Low (15%)

NA 28.51 % SVS 9.95% LYS 61.54% SLS 24.71 % SVS 8.63% LYS 66.67% SLS 28.51 % SVS 9.95% LYS 61.54% SLS

NA: not applicable.

It was observed that no dry powder could be produced by spray drying SVS-LYS from 5% Tween20 aqueous solution, which was not surprising considering that Tween20 is a viscous liquid at room temperature. Using SLS solutions, different formulation (0.5% or 5% SLS solution) and processing conditions (Tinlet , and Toutlet which was an outcome of Tinlet and the pump rate) were applied (Table 2). As can be seen from Table 2, higher Tinlet , and consequently Toutlet , temperatures produced low yields, whereas lower temperatures improved the results when 5% SLS was used. Using more dilute SLS (0.5%) improved the yield at inlet 110 ˝ C, but reducing the volume of the 5% SLS solution and decreasing the inlet to 100 ˝ C improved the yield slightly further. Thus, the best spay drying conditions were the ones where Tinlet was 100 ˝ C, Toutlet was 45 ˝ C, pump rate was 3.9 mL/min when using a 5% SLS solution. These conditions were used to produce spray-dried SVS-LYS for further studies. Spray drying of drugs, having a relatively low glass transition temperature (Tg ), such as SVS, is difficult since often stable amorphous products in the form of a free flowing powder cannot be obtained [21]. Toutlet being above the Tg values of the products may be the reason for the relatively low yields, making the powder stick to the surfaces of the spray dryer [27]. The solid state form (amorphous/crystalline) and the Tg -values of the powders obtained were subsequently studied. 2.3. Characterization of the Spray-Dried Materials 2.3.1. X-ray Powder Diffraction (XRPD) The XRPD diffractograms of crystalline SLS and the spray-dried formulations are shown in Figure 2. In addition, the diffractogram of SVS-LYS CM (cryo-milled) physically mixed with SLS is shown (SVS-LYS CM + SLS). The XRPD-analysis of the powder samples (Figure 2) revealed that pure SLS showed characteristic diffraction peaks [28] at 2θ values of 6.8˝ , 18.2˝ , 20.3˝ , 20.7˝ , 21.8˝ , and 31.7˝ and that the remaining crystallinity in the spray-dried samples originated from SLS, i.e., no SVS and/or LYS diffraction peaks [7] were observed. However, some changes in SLS crystallinity in the spray-dried samples might have occurred. The SLS peak at a 2θ-value of 6.8˝ had disappeared completely in SVS-LYS spray-dried from 5% SLS and in SVS spray-dried from 5% SLS. The SLS peak at a 2θ-value of 31.7˝ had disappeared in all spray-dried samples. The major diffractions at a 2θ-value of

21535

Molecules 2015, 20, 21532–21548

approx. 21˝ had remained unchanged in SVS-LYS spray-dried from 0.5% SLS and in SVS spray-dried from 5% SLS. However, the shape of this diffraction feature had changed in SVS-LYS from 5% SLS, indicating a possible change in the crystal structure of SLS. Molecules 2015, 20, page–page

Figure 2. XRPD diffractograms of SVS, LYS, SLS, and the spray-dried formulations of SVS-LYS from

Figure 2. XRPD diffractograms of SVS, LYS, SLS, and the spray-dried formulations of SVS-LYS from 0.5% SLS, SVS-LYS from 5% SLS, SVS from 0.5% SLS, and SVS-LYS CM + SLS. 0.5% SLS, SVS-LYS from 5% SLS, SVS from 0.5% SLS, and SVS-LYS CM + SLS. 2.3.2. Particle Surface Characteristics ElementalCharacteristics analysis of the spray-dried particle surfaces was conducted with a scanning electron 2.3.2. Particle Surface microscope (SEM) equipped with an energy-dispersive X-ray (EDS) detector. As can be seen from

Elemental of the particle was(H). conducted a scanning Figure analysis 1, SVS consists onlyspray-dried of carbon (C), oxygen (O),surfaces and hydrogen In additionwith to these elements, electron LYS also contains nitrogen (N). an SLS energy-dispersive in turn contains C, O, H,X-ray sulfur (S), and sodium (Na). As Based on this, microscope (SEM) equipped with (EDS) detector. can be seen from SLS could be differentiated from the other components in the particles. Since SLS is a surface-active agent, Figure 1, SVS consists only of carbon (C), oxygen (O), and hydrogen (H). In addition to these elements, it could be anticipated that it would accumulate on the droplet surfaces in the atomized feed solution and LYS also contains nitrogen (N). SLS in turn contains C, O, H, sulfur (S), and sodium (Na). Based thus on dry particle surfaces during the spray drying process [29]. This has also been observed when on this, SLS could beis differentiated from thesolution other above components in theconcentration particles. (CMC) Since SLS is a the surfactant present in the spray drying its critical micelle [30], as is the case in the SLS used inthat this study. Although forming micelles above the CMC, surface-active agent, it could besolutions anticipated it would accumulate on the droplet surfaces in the gas–liquid interfaceand of the droplets will be saturatedsurfaces with the surfactant [31], which then the atomized feed solution thus on dry particle during monomers the spray drying process [29]. accumulate on the dry particle surfaces. The EDS measurements revealed that SLS in fact coated the This has also been observed when the surfactant is present in the spray drying solution above its particles as Na and S were concentrated on the particle surfaces (Figure 3). However, there are differences critical micelle concentration is the case in thetheSLS solutions usedWhen in this study. in the relative amounts of(CMC) Na and S [30], on the as surface when comparing different formulations. spray-dried from more dilutethe SLSgas–liquid solution (i.e.,interface 0.5%), there to be less Na and Although SVS-LYS formingwas micelles above the CMC, ofseems the droplets will beS saturated (i.e., less SLS) on the particle surface compared to the formulations dried from the 5% SLS solution with the surfactant monomers [31], which then accumulate on the dry particle surfaces. The EDS (even though the absolute amount of SLS in the final powder was the same). measurements revealed that SLS in fact coated the particles as Na and S were concentrated on the particle surfaces (Figure 3). However, there are differences in the relative amounts of Na and S on the surface when comparing the different formulations. When SVS-LYS was spray-dried from more dilute SLS solution (i.e., 0.5%), there seems to be less Na and S (i.e., less SLS) on the particle surface compared to the formulations dried from the 5% SLS solution (even though the absolute amount of SLS in the final powder was the same). 2.3.3. Differential Scanning Calorimetry (DSC) The thermal properties of the prepared samples were investigated using DSC. Table 3 shows the 5 melting (Tm ), glass transition (Tg ), and recrystallization temperatures (Trc ) of all samples.

21536

Molecules 2015, 20, 21532–21548 Molecules 2015, 20, page–page Molecules 2015, 20, page–page  Molecules 2015, 20, page–page 

   Figure 3. Distribution of Na and SS (•)  on the spray-dried particle surfaces measured by EDS: Distribution ((•) ) and  ((•) onthe  thespray‐dried  spray-driedparticle  particle surfaces  surfaces measured  Figure  Figure 3.  3. Distribution  Distribution of  of Na  Na (•)  (•)  and S  (•) )on  on  the  spray‐dried  particle  surfaces  measured by  by EDS:  EDS:     SVS-LYS from from 5% 5% SLS; SLS; (b) (b) SVS-LYS SVS-LYSfrom from0.5% 0.5%SLS; SLS;and and(c) (c)SVS SVSfrom from5% 5%SLS. SLS. (a) SVS-LYS (a) SVS‐LYS from 5% SLS; (b) SVS‐LYS from 0.5% SLS; and (c) SVS from 5% SLS.  (a) SVS‐LYS from 5% SLS; (b) SVS‐LYS from 0.5% SLS; and (c) SVS from 5% SLS. 

2.3.3. Differential Scanning Calorimetry Table 3. Melting temperature (Tm ), glass(DSC) transition (Tg ), and recrystallization temperature (Trc ) of the 2.3.3. Differential Scanning Calorimetry (DSC)  2.3.3. Differential Scanning Calorimetry (DSC)  pure starting materials compared to the prepared samples.

The thermal properties of the prepared samples were investigated using DSC. Table 3 shows the The thermal properties of the prepared samples were investigated using DSC. Table 3 shows the  The thermal properties of the prepared samples were investigated using DSC. Table 3 shows the  melting (T m ),), glass transition (T glass transition (Tgg˝g), and recrystallization temperatures (T ), and recrystallization temperatures (Trcrcrc) of all samples.  ) of all ˝samples. melting (T melting (T m m), glass transition (T Material Tg ( ), and recrystallization temperatures (T C) Trc (˝ C) Tm SVS (˝ C) T) of all samples.  Tm SLS (˝ C) m LYS ( C) SVS a

29.0 ˘ 0.6

NA

139.9 ˘ 0.2

NA

NA

107.0 ˘ 1.2 T rc (°C) 73.1 0.8 T Trc  rc ˘ (°C) (°C) 68.6 ˘ 9.6 NA NA  NA  ND

134.1 ˘ 0.1

199.9 ˘ 3.2

NA

Table 3. Melting temperature (T m), glass transition (Tggg), and recrystallization temperature (T ),), and recrystallization temperature (T and recrystallization temperature (Trcrcrc) of the  )) of the  of the Table 3. Melting temperature (T Table 3. Melting temperature (T ), glass transition (T LYS a NA NA 212.4 ˘ 0.5 a NA 68 b mm), glass transition (T pure starting materials compared SLS ND to the prepared NA samples. NA NA 194.3 ˘ 0.5 c pure starting materials compared to the prepared samples.  pure starting materials compared to the prepared samples. 

SVS-LYS CM a Material SVS-LYS from 0.5% SLS Material  Material  SVS-LYS from 5% aSLS a a  SVS SVS  SVS from SVS  5% SLS

33.2 ˘ 0.9 T (°C) 23.6 ˘ T Tgg5.7 g (°C)  (°C) 26.2 ˘ 3.8 29.0 ± 0.6 29.0 ± 0.6 13.729.0 ± 0.6 ˘ 0.6

T SVS T LYS T m SLS (°C) ˘(°C) 0.2 ND(°C) T 162.9 ˘ 0.4 T Tmmm143.7 SVS (°C) SVS (°C) T TmmmLYS (°C)  LYS (°C)  Tm  m SLS (°C) SLS (°C) 140.6 ˘1.0 ND 157.6 ˘ 0.4     139.9 ± 0.2 NA NA 139.9 ± 0.2  139.9 ± 0.2  NA  NA  NA NA˘ 0.6 ND NA 182.7 aaa    bbb    c LYS 68 NA NA 212.4 ± 0.5 a aa  NA LYS  LYS  68  68  NA  NA  NA  NA  212.4 ± 0.5  212.4 ± 0.5  NA  NA  a : Values b from [7]; : Value from [32]; : Another endothem was seen at 103.1 ˘ 7.2; NA: not applicable;     SLS ND NA NA NA 194.3 ± 0.5 ccc ND: not detected. SLS  SLS  ND ND NA  NA  NA  NA  NA  NA  194.3 ± 0.5  194.3 ± 0.5  SVS-LYS CM a aa  33.2 ± 0.9 107.0 ± 1.2 107.0 ± 1.2 134.1 ± 0.1  134.1 ± 0.1 199.9 ± 3.2 NA SVS‐LYS CM  SVS‐LYS CM  33.2 ± 0.9 33.2 ± 0.9 107.0 ± 1.2 134.1 ± 0.1  199.9 ± 3.2  199.9 ± 3.2  NA  NA  SVS-LYS from 0.5% 23.6 ± 5.7 73.1 ± 0.8 143.7 0.2 observed ND 162.9 ± 0.4 of SVS‐LYS from 0.5% SLS  SVS‐LYS from 0.5% SLS  23.6 ± 5.7 23.6 ± 5.7 73.1 ± 0.8 143.7 ± 0.2  143.7 ± 0.2  ND  ND in the 162.9 ± 0.4  162.9 ± 0.4  From Table 3 it canSLS be seen that two 73.1 ± 0.8 endothermic peaks±were thermogram ˝ ˝ SVS-LYS from 5% SLS 26.2 ± 3.8 68.6 ± 9.6 140.6 ±1.0 ND 157.6 0.4 SVS‐LYS from 5% SLS  26.2 ± 3.8 68.6 ± 9.6 140.6 ±1.0  140.6 ±1.0  ND  157.6 ± 0.4  pure SVS‐LYS from 5% SLS  SLS at 103.13 C (loss of 26.2 ± 3.8 water) and 68.6 ± 9.6 194.25 C (melting), which is inND  agreement157.6 ± 0.4  with ±literature SVS from 5% SLS 13.7 ± 0.6 ND ND NA 182.7 ± 0.6 SVS from 5% SLS  SVS from 5% SLS  13.7 ± 0.6 13.7 ± 0.6 ND  ND  ND  NA  NA  samples. 182.7 ± 0.6  182.7 ± 0.6  data [33–36]. The lower endotherm of SLS was observed inND  all the spray-dried In addition, cc: Another endothem was seen at 103.1 aaa: Values from [7];  bbb: Value from [32];  :: Values from [7];  Values from [7]; :: Value from [32];  Value from [32]; cat ± 7.2; NA: not applicable; a recrystallization peak was observed a temperature range of 68–75˝ C for the spray-dried samples : Another endothem was seen at 103.1 ± 7.2; NA: not applicable;  : Another endothem was seen at 103.1 ± 7.2; NA: not applicable;  ND: not detected. containing LYS, which was in accordance with the Trc previously measured for amorphous SVS [7]. ND: not detected.  ND: not detected.  Previously, it has been found that the Trc was increased to 107 ˝ C in the case of SVS-LYS CM, pointing From 3 it can be seen that two endothermic peaks were observed the thermogram of From Table 3 it can be seen that two endothermic peaks were observed in the thermogram of  From Table 3 it can be seen that two endothermic peaks were observed in the thermogram of  towards a Table stabilizing effect by formation of the co-amorphous mixture withinLYS [7]. A similar pure SLS at 103.13 °C (loss of water) and 194.25°C (melting), which is in agreement with literature pure SLS at 103.13 °C (loss of water) and 194.25°C (melting), which is in agreement with literature  pure SLS at 103.13 °C (loss of water) and 194.25°C (melting), which is in agreement with literature  phenomenon, however, was not observed when SLS was included in the spray-dried formulations data [33–36]. lower endotherm of SLS was in all the spray-dried samples. In addition, data [33–36]. The lower endotherm of SLS was observed in all the spray‐dried samples. In addition,  data [33–36]. The lower endotherm of SLS was observed in all the spray‐dried samples. In addition,  with SVS andThe LYS. After recrystallization, an observed endothermic melting peak of SVS was observed at a recrystallization peak was observed at a temperature range of 68–75°C for the spray-dried samples ˝ a recrystallization peak was observed at a temperature range of 68–75°C for the spray‐dried samples  a recrystallization peak was observed at a temperature range of 68–75°C for the spray‐dried samples  approx. 140 C in these samples. In contrast, neither a recrystallization peak nor a melting peak containing whichinwas accordanceofwith Trcrcrc5% previously measured for amorphous SVS [7]. containing LYS, which was in accordance with the T containing LYS, which was in accordance with the T  previously measured for amorphous SVS [7].   previously measured for amorphous SVS [7].  of SVS wereLYS, observed the in thermogram SVS the from SLS, which is indicative of a higher stability Previously, it has been found that the T rc was increased to 107 °C in the case of SVS-LYS CM, pointing Previously, it has been found that the T Previously, it has been found that the T rcrc was increased to 107 °C in the case of SVS‐LYS CM, pointing   was increased to 107 °C in the case of SVS‐LYS CM, pointing  against recrystallization for this formulation [37,38]. Instead, an SLS melting peak was observed, but towards aa stabilizing  stabilizing effect by formation of the co-amorphous mixture with LYS [7]. A similar towards  towards a  stabilizing effect  effect by  by formation  formation of  of the  the co‐amorphous  co‐amorphous mixture  mixture with  with LYS  LYS [7].  [7]. A  A similar  similar  phenomenon, however, was not observed when SLS was included in the spray-dried formulations phenomenon, however, was not observed when SLS was included in the spray‐dried formulations  phenomenon, however, was not observed when SLS was included in the spray‐dried formulations  21537 666

Molecules 2015, 20, 21532–21548

at approximately 10 degrees lower than in pure SLS. In the formulations containing LYS, the SLS melting peak was very small and observed at temperatures 30 degrees lower than in pure SLS. This may be an indication of changes in the crystalline character of SLS and/or for partial miscibility of SLS with SVS and LYS. Furthermore, one single Tg was observed for all spray-dried samples, indicating formation of a homogenous amorphous phase [39,40]. In general, the observed Tg was somewhat lower than for pure amorphous SVS (Table 1) and amorphous SVS-LYS (determined previously to be 33.2 ˝ C [7]). This indicated that SLS, which has a low melting point and thus a low Tg [41], was partially miscible with the amorphous phase formed by SVS and LYS and acted as a plasticizer in the mixtures. SLS seemed to have the largest plasticizing effect on SVS from 5% SLS, as the Tg of this mixture was found to be lowest. After these measurements SVS-LYS from 0.5% SLS was not characterized further due to a long spray drying time as a consequence of the large volume of 0.5% SLS solution required for dissolving SVS. 2.3.4. FTIR After the spray drying process with SLS, peak broadening and peak shifts were seen at the SVS ester C=O stretch absorption region from 1695 cm´1 to 1715 cm´1 and from 3550 to 3450 cm´1 (SVS OH-region) when compared to crystalline SVS (Figure 4). These shifts were also observed in amorphous SVS and co-amorphous SVS-LYS and they can be considered to originate from the amorphization of SVS [7]. The most intense band observed in the spectra of the spray-dried samples was the SO3 asymmetric vibrational feature (νas (SO3 )) of SLS, which in general is observed as a double peak in crystalline, ordered SLS [42–44]. In the spectrum of crystalline SLS, the SLS SO3 asymmetric stretching is a large duplet at 1244 cm´1 and 1217 cm´1 with a small shoulder at 1180 cm´1 (Figure 4). It covers the SVS-related absorption feature at 1245 cm´1 (seen in amorphous SVS and SVS-LYS CM). This SLS SO3 -absorption feature was broadened in SVS from 5% SLS but merged to one large peak between 1170 cm´1 and 1300 cm´1 in SVS-LYS from 5% SLS, indicating disruption of the structural order of SLS. In addition, the shoulder peak at 1180 cm´1 was not seen in both types of spray-dried mixtures (with or without LYS), which may be indicative of interactions of this group with SVS. Peak broadening of other SLS-related features was observed in the spray-dried samples. The FTIR spectra of SVS-LYS from 5% SLS and the corresponding physical mixture (PM) showed two peaks at 1580 cm´1 and 1512 cm´1 (amide I and amide II bands of LYS). These peaks were less sharp in SVS-LYS from 5% SLS than in the corresponding PM, but they did not merge into one large peak upon spray drying, in contrast to what was observed in co-amorphous SVS-LYS CM [7]. The aliphatic chain vibrations of LYS, which shifted to lower wavelengths upon formation of co-amorphous SVS´LYS (i.e., to 1350 cm´1 and 1312 cm´1 ) [7], can be seen at even lower wavelengths in SVS-LYS from 5% SLS (i.e., at 1348 cm´1 and 1309 cm´1 ). In addition, the absorption band at 1390 cm´1 (originating from LYS) appears less sharp in the spectra of SVS-LYS from 5% SLS than in the spectrum of SVS-LYS CM. These changes indicate that in SVS-LYS from 5% SLS, SLS may have disturbed formation of the weak interactions between SVS and LYS, which formed in co-amorphous SVS-LYS. This is also in accordance with the DSC results, which showed that an increase in Trc in SVS-LYS CM compared to amorphous SVS was not observed when SLS was incorporated in the formulation. 2.3.5. Dissolution Properties In addition to the spray-dried formulations, the dissolution profiles of the corresponding physical mixtures and SVS-LYS CM + SLS were determined in phosphate buffer (pH 7.2) (Figure 5). This was done to have a fully crystalline reference (PM), a reference with amorphous SVS-LYS (CM) physically mixed with SLS, and a mixture where SVS-LYS and SLS were all processed together.

21538

Molecules Molecules 2015, 2015,20, 20,21532–21548 page–page

Figure 4. FTIR spectra of the prepared samples and corresponding physical mixtures (PMs). In addition, the spectra of amorphous SVS (prepared by CM) and SVS-LYS CM are shown for comparison.

2.3.5. Dissolution Properties In addition to the spray-dried formulations, the dissolution profiles of the corresponding physical mixtures and SVS-LYS CM + SLS were determined in phosphate buffer (pH 7.2) (Figure 5). This was done to have a fully crystalline reference (PM), a reference with amorphous SVS-LYS (CM) physically mixed with SLS, and a mixture where SVS-LYS and SLS were all processed together. Figure 4. FTIR spectra of the prepared samples and corresponding physical mixtures (PMs). Figure 4. FTIR spectra of the prepared samples and corresponding physical mixtures (PMs). In addition, In addition, the spectra of amorphous SVS (prepared by CM) and SVS-LYS CM are shown the spectra of amorphous SVS (prepared by CM) and SVS-LYS CM are shown for comparison. for comparison.

2.3.5. Dissolution Properties In addition to the spray-dried formulations, the dissolution profiles of the corresponding physical mixtures and SVS-LYS CM + SLS were determined in phosphate buffer (pH 7.2) (Figure 5). This was done to have a fully crystalline reference (PM), a reference with amorphous SVS-LYS (CM) physically mixed with SLS, and a mixture where SVS-LYS and SLS were all processed together.

Figure5.5. The The cumulative cumulative amounts andand (b) (b) SVSSVS Figure amounts ofof SVS SVSdissolved dissolvedfor for(a)(a)SVS-LYS SVS-LYSformulations formulations formulations in phosphate buffer (pH 7.2). The point of 24 h is not shown, since it was not statistically formulations in phosphate buffer (pH 7.2). The point of 24 h is not shown, since it was not statistically differentfrom fromthe thevalue value obtained obtained at at 480 480 min. SVS-LYS PM, and SVS-LYS CMCM different min. The Thedata dataforforSVS, SVS, SVS-LYS PM, and SVS-LYS formulations is taken from [45]. formulations is taken from [45].

The saturation dissolvable amount of crystalline SVS (without SLS) in these conditions has been The saturation dissolvable amount of crystalline SVS (without SLS) in these conditions has been previously determined to be 0.15 mg [45]. The same study showed that amorphization of SVS alone previously to be 0.15 mg [45]. Thedissolved same study thatformulations amorphization SVS Figure determined 5. The cumulative amounts of SVS for showed (a) SVS-LYS andof(b) SVSalone did not improve in the dissolution of(pH SVS. InThe contrast, combining SVS with LYS in a co-amorphous formulations phosphate buffer 7.2). point 8 of 24 h is not shown, since it was not statistically formulation (SVS-LYS CM) improved the dissolution significantly (the concentrations were different from the value obtained at 480 min. The data for SVS, SVS-LYS PM, and SVS-LYS CMthree timesformulations higher withisSVS-LYS CM than with SVS-LYS PM). LYS in a simple crystalline physical mixture taken from [45]. with SVS was not able to enhance the dissolution compared to crystalline SVS (see Figure 5a) [45]. In The amount crystalline SVS (without SLS) in these conditions has been Figure 5 itsaturation can also bedissolvable seen that SLS had a of significant solubilizing effect (approx. 10-fold) on crystalline previously determineddissolved to be 0.15amount mg [45].ofThe thatof amorphization of SVS alone SVS, as the maximum SVSsame was study 1.8 mgshowed in the case PMs. However, when the did not improve theCM dissolution of SVS.mixed In contrast, combining SVS with LYS ineven a co-amorphous amorphous SVS-LYS was physically with SLS, the dissolution increased further (this formulation had (SVS-LYS CM) improved the dissolution (the concentrations were three combination the best dissolution properties amongstsignificantly the ternary formulations). The spray-dried times higher(SVS-LYS with SVS-LYS thanhad withdissolution SVS-LYS PM). LYS inalmost a simple crystalline physical mixture formulation from CM 5%SLS) properties identical to SVS-LYS-SLS PM with SVS was not able to enhance the dissolution compared to crystalline SVS (see Figure 5a) [45].

21539 8

Molecules 2015, 20, 21532–21548

(not significantly different (p > 0.05) from the corresponding PM). However, in the case of SVS-LYS from 5% SLS (Figure 5a), the maximum released amount of SVS (1.8 mg) was already reached after 90 min while the maximum released amount of SVS (1.8 mg) for SVS-LYS-5% SLS PM was seen later, after approximately 360 min. It can be stated that the dissolution curve of SVS-LYS from 5% SLS was not significantly different (p > 0.05) from the corresponding PM. Thus, the amorphous form in this case did not enhance the dissolution further. However, the dissolution of SVS-LYS CM + SLS was significantly faster (p < 0.05) than that of SVS-LYS spay dried from 5% SLS. In addition this formulation enabled higher concentrations of SVS than the others. The maximum released amount of SVS (2.46 mg) in the case of SVS-LYS CM + SLS was already seen after 30 min. This formulation was thus able to provide a “spring and parachute” effect [46], i.e., rapid supersaturation (peak concentration at 30 min), which was maintained until approx. 240 min. Compared to the dissolution properties of SVS-LYS CM determined previously in similar conditions (shown in Figure 5a) [45], the amount of dissolved SVS was approx. 10-fold with SVS-LYS CM + SLS. In addition, the peak concentration was achieved much faster with SVS-LYS CM + SLS (i.e., 30 min vs. 480 min). In contrast, SVS-SLS PM (Figure 5b) had similar dissolution properties to SVS-LYS from 5%SLS and SVS-LYS-SLS PM. Thus, it seems that LYS does not have a positive effect on dissolution in a completely crystalline mixture (SVS-LYS-SLS PM vs. SVS-SLS PM) or when spray dried as a ternary combination (SVS-LYS from 5%SLS vs. SVS-LYS-SLS PM vs. SVS-SLS PM). LYS is able to improve the dissolution of SVS only when combined with SVS as a co-amorphous binary formulation (SVS-LYS CM or SVS-LYS CM + SLS). An even more enhanced “spring and parachute” effect than that seen with SVS-LYS CM + SLS was observed for SVS spray-dried from 5% SLS (Figure 5b), with the maximum dissolved amount of SVS (3.23 mg) reached after 90 min. This was almost twice the maximum dissolved amount of the PM mixture (Figure 5). The supersaturated state generated by the amorphous form was maintained throughout the experiment, as the dissolution from the spray-dried formulation was significantly enhanced (p < 0.05) compared to that of the corresponding PM and to the second best formulation, i.e., SVS-LYS CM + SLS. The results thus indicate that the dissolution properties of SVS from the co-amorphous SVS-LYS can be improved by physically mixing with SLS but co-processing of these three components together was not beneficial. Instead, the best dissolution properties for SVS dissolution were achieved by spray drying SVS with SLS. 2.3.6. Physical Stability The physical stability of the formulations was investigated after storage under different conditions i.e., 4 ˝ C/0% RH, 40 ˝ C/0% RH, and 25 ˝ C/60% RH. The stability of the spray-dried formulations was compared with SVS-LYS CM + SLS. This was done in order to investigate whether there was any difference in the stability if SLS is added simply as a crystalline component to the amorphous SVS-LYS CM in contrast to all components being processed together. Interestingly, even though the formulations were only partially amorphous initially (remaining SLS crystallinity) the changes observed as a function of time in different conditions were relatively small (Figure 6). The SVS-LYS formulation spray-dried from 5% SLS (Figure 6a), was found to be stable at 4 ˝ C/0% RH and 40 ˝ C/0% RH for at least one year. Compared to the freshly prepared sample (Figure 2), the only changes occurring under these conditions were changes in SLS-related diffractions. The SLS-related peak at 6.8˝ 2θ appeared after five weeks of storage at 4 ˝ C/0% RH and after two months at 40 ˝ C/0% RH. No peaks related to crystalline SVS or LYS appeared during the storage. When stored at 25 ˝ C/60% RH, peaks appeared also at 9.5˝ (at five weeks) and at 22.5˝ 2θ (at eight months) which can be assigned to SVS [8]. Interestingly, the peak shapes of the four major SLS-diffractions (between 20˝ and 22˝ 2θ) changed during storage, depending on the conditions. At 4 ˝ C/0% RH, the relative height of the third peak decreased and at 25 ˝ C/60% RH and 40 ˝ C/0% RH it disappeared completely. Thus, only samples stored at 25 ˝ C/60% RH for seven months and

21540

Molecules 2015, 2015, 20, 20, 21532–21548 page–page Molecules

the third peak decreased and at 25 °C/60% RH and 40 °C/0% RH it disappeared completely. Thus, ˝ C/0% RH for 12 months showed a diffraction pattern that resembles the diffraction pattern of at 40 samples only stored at 25 °C/60% RH for seven months and at 40 °C/0% RH for 12 months showed a pure crystalline SLS. diffraction pattern that resembles the diffraction pattern of pure crystalline SLS.

Figure 6. X-ray diffractograms of the spray-dried powders after storage under different conditions: SVS-LYS spay spay dried dried from from5% 5%SLS; SLS;(b) (b)SVS SVSspray-dried spray-driedfrom from5% 5%SLS; SLS;(c) (c)SVS-LYS SVS-LYSCM CM++SLS. SLS. (a) SVS-LYS

In the the case case of of SVS SVS spray spray dried dried from from 5% 5% SLS SLS (Figure (Figure 6c), 6c), the the observable observable changes changes in in all all conditions conditions In were related At 4At °C/0% RH, the SLS peaks 6.8° and 18.2° 2θ appeared three ˝ and were related to toSLS SLSdiffractions. diffractions. 4 ˝ C/0% RH, the SLS atpeaks at 6.8 18.2˝ 2θ after appeared weeks of storage but no other changes occurred in the diffractogram during a one-year storage period after three weeks of storage but no other changes occurred in the diffractogram during a one-year (the major diffractions were more or less already in thealready fresh sample whensample compared to storage period (the major diffractions wereunchanged more or less unchanged in the fresh when crystalline to SLS). At 25 °C/60% and˝ C/60% 40 °C/0% RH, these two weeks of compared crystalline SLS). RH At 25 RH and 40 ˝peaks C/0%appeared RH, thesealready peaks after appeared already storage. In addition, a SLS peak at 31.7° 2θ was seen at all storage conditions. However, no other ˝ after two weeks of storage. In addition, a SLS peak at 31.7 2θ was seen at all storage conditions. changes were observed in thewere diffractograms year of storage (40one °C/0% and seven ˝ C/0% However, no other changes observed inuntil the one diffractograms until yearRH) of storage (40 months of storage (25 °C/60% RH). ˝ RH) and seven months of storage (25 C/60% RH). In the the case case of of SVS-LYS SVS-LYS CM CM ++ SLS, SLS, two two new new peaks peaks appeared appeared during during the the storage storage period period (at (at seven seven In months at RHRH andand at eight months in thein other As crystalline SLS was physically ˝ C/0% months at 40 40°C/0% at eight months the conditions). other conditions). As crystalline SLS was mixed with co-amorphous SVS-LYS, the changes were related to recrystallization of this co-amorphous physically mixed with co-amorphous SVS-LYS, the changes were related to recrystallization of this mixture. The peaks appearing at 7.8° 2θ (at eight months in eight 4 °C/0% RH and months in 40 °C/ ˝ 2θ (at ˝ C/0% co-amorphous mixture. The peaks appearing at 7.8 months in 4seven RH and seven 0% RH) in and (atRH) eight months in 4eight °C/0% RH and °C/60% RH , and sevenRH months °C/ ˝ C/0% months 40 17.2° and 17.2˝ (at months in 425˝ C/0% RH and 25 ˝at C/60% , and in at 40 seven 0% RH) could be attributed to recrystallization of SVS from the mixture [7,8]. ˝ months in 40 C/0% RH) could be attributed to recrystallization of SVS from the mixture [7,8]. The FTIR-spectra FTIR-spectraofof stored samples are shown in 7.Figure Inofthe case of SVS-LYS The thethe stored samples are shown in Figure In the 7. case SVS-LYS spray-dried spray-dried 5% SLS, the SO3 asymmetric vibrational double peak of SLS that was merged upon from 5% SLS,from the SO 3 asymmetric vibrational double peak of SLS that was merged upon spray drying −1–1300 cm−1) was again split to a double peak during storage. The peak spray drying (at 1170 cm ´ 1 ´ 1 (at 1170 cm –1300 cm ) was again split to a double peak during storage. The peak splitting splitting wasafter visible aftermonths eight months when stored 4 °C/0%RH °C/0%RH.Instead, Instead,at at storage storage ˝ C/0%RH. was visible eight when stored at 4 ˝at C/0%RH andand 40 40 conditions of 25 °C/60%RH the peak splitting was seen at nine weeks. In addition, the shoulder peak at ˝ conditions of 25 C/60%RH the peak splitting was seen at nine weeks. In addition, the shoulder peak −1 (from SLS), which disappeared after spray drying, reappeared in its original position at 1180 cm at 1180 cm´1 (from SLS), which disappeared after spray drying, reappeared in its original position at week one one in in the the FTIR FTIR spectrum spectrum of ofthe thesample samplestored storedat at25 25˝°C/60%RH. other storage storage conditions, conditions, week C/60%RH. In In the the other this peak did not reappear. However, the SVS ester C=O stretch absorption band, which shifted from this peak did not reappear. However, the SVS ester C=O stretch absorption band, which shifted from 1695 cm−1 to 1715 cm−1 upon amorphization, was unchanged at all storage conditions. Instead, the 10 21541

Molecules 2015, 20, 21532–21548

Molecules 2015, 20, page–page

1695 cm´1 to 1715 cm´1 upon amorphization, was unchanged at all storage conditions. Instead, the amide II and and amide amide IIII bands bands of of LYS LYSatat1580 1580cm cm and 1512 1512 cm cm´−11 became ´−1 1 and amide became sharper sharper upon upon storage storage at at all conditions. all conditions.

Figure 7.7.FTIR-spectra of theofspray-dried powders after storage different SVS-LYS Figure FTIR-spectra the spray-dried powders afterunder storage underconditions: different (a) conditions: spray-dried 5% SLS; (b) SVS spray-dried 5% SLS;from (c) SVS spray-dried 5% SLSfrom and (a) SVS-LYS from spray-dried from 5% SLS; (b) SVS from spray-dried 5% SLS; (c) SVS from spray-dried SVS-LYS CM + 5% SLS. 5% SLS and SVS-LYS CM + 5% SLS.

In SVS spray-dried from 5% SLS, no changes were observed in the SO3 asymmetric vibrational In SVS spray-dried from 5% SLS, no changes were observed in the SO3 asymmetric vibrational double peak of SLS compared to the fresh sample during storage. The shoulder peak at 1180 cm−1 also double peak of SLS compared to the fresh sample during storage. The shoulder peak at 1180 cm´1 reappeared in the FTIR spectrum of the sample stored at 25 °C/60%RH at seven months. However, also reappeared in the FTIR spectrum of the sample stored at 25 ˝ C/60%RH at seven months. the SVS ester C=O stretch absorption band at 1715 cm−1 was unchanged in all storage conditions However, the SVS ester C=O stretch absorption band at 1715 cm´1 was unchanged in all storage (except broadened in 25 °C/60%RH). The spectrum of the sample stored at 25 °C/60%RH showed a conditions (except broadened in 25 ˝ C/60%RH). The spectrum of the sample stored at 25 ˝ C/60%RH rise in the baseline level, which probably indicates moisture absorption into the sample. showed a rise in the baseline level, which probably indicates moisture absorption into the sample. In SVS-LYS CM + SLS the SLS-shoulder peak at 1180 cm−1 became higher at 25 °C/60%RH at In SVS-LYS CM + SLS the SLS-shoulder peak at 1180 cm´1 became higher at 25 ˝ C/60%RH seven months. In addition, in these conditions the SVS ester C=O stretch absorption band at 1715 cm−1 at seven months. In addition, in these conditions the SVS ester C=O stretch absorption band at was slightly split, indicating recrystallization of SVS. 1715 cm´1 was slightly split, indicating recrystallization of SVS. In a previous study with the co-amorphous SVS-LYS CM formulation [7], it was found that this In a previous study with the co-amorphous SVS-LYS CM formulation [7], it was found that this formulation was still X-ray amorphous after five months of storage at 4 ˝°C/0% RH. At an elevated formulation was still X-ray amorphous after five months of storage at 4 C/0% RH. At an elevated temperature (40 °C/0%RH), the mixture showed signs of recrystallization within three months and at temperature (40 ˝ C/0%RH), the mixture showed signs of recrystallization within three months and at elevated humidity (25 °C/60% RH) at day 56. Compared to this, all the formulations prepared in the elevated humidity (25 ˝ C/60% RH) at day 56. Compared to this, all the formulations prepared in the current study were able to significantly extend the stability of amorphous SVS, which was surprising current study were able to significantly extend the stability of amorphous SVS, which was surprising due to the fact that at least partially crystalline excipient (SLS) was present in the formulations. due to the fact that at least partially crystalline excipient (SLS) was present in the formulations. In addition, surfactants cannot be considered as stabilizing excipients for amorphous materials in In addition, surfactants cannot be considered as stabilizing excipients for amorphous materials in general [47]. It has been found that surfactants can increase the crystal growth rate of amorphous general [47]. It has been found that surfactants can increase the crystal growth rate of amorphous celecoxib if they are miscible with the drug [41]. However, SLS, being immiscible with celecoxib, was celecoxib if they are miscible with the drug [41]. However, SLS, being immiscible with celecoxib, found to have very little impact on the growth rate of celecoxib crystals. When the miscibility was was found to have very little impact on the growth rate of celecoxib crystals. When the miscibility improved by adding polyvinylpyrrolidone (PVP) to form a ternary dispersion of celecoxib, SLS was then found to increase the growth rate of celecoxib compared to the corresponding binary celecoxib– PVP blends. In the current study, the best stabilizing 21542 effect was found (at least 12 months in dry 11

Molecules 2015, 20, 21532–21548

was improved by adding polyvinylpyrrolidone (PVP) to form a ternary dispersion of celecoxib, SLS was then found to increase the growth rate of celecoxib compared to the corresponding binary celecoxib–PVP blends. In the current study, the best stabilizing effect was found (at least 12 months in dry conditions), when SLS was spray-dried with SVS (and LYS) instead of being physically mixed with the co-amorphous SVS-LYS. Since SLS was only partially miscible with SVS (i.e., lowered the Tg of the mixture slightly), and thus remained mainly in the crystalline phase in the studied mixtures, it seems less likely that miscibility of SLS had an impact on the stability of SVS. Instead, the stabilizing effect could be due to a protective coating produced by SLS upon spray drying, which provided physical separation of the amorphous material (Figure 3) [48]. Furthermore, the hygroscopicity of SLS has been found to be low below RH values of 75%, after which it shows deliquescence [49]. Thus, SLS might protect the amorphous content form ambient moisture below this RH value. The results also indicated that in this case incorporating LYS as a co-amorphous former did not bring significant stability or dissolution advantages when compared to formulations containing only amorphous SVS. 3. Experimental Section 3.1. Materials Simvastatin (SVS) and the amino acid lysine (LYS) were supplied by Hangzou Dayangchem Co., Ltd. (Hangzou, China). The solubilizers polysorbate 20 (Tween20), polyvinylpyrrolidone K-30 (PVP), and Pluronic F-68 were obtained from Sigma-Aldrich Chemie GmbH (Steinheim, Germany). Sodium lauryl sulfate (SLS) and Soluplus were provided by YA-Kemia Oy (Helsinki, Finland) and BASF SE (Ludwigshafen, Germany) respectively. The compounds were used as received. 3.2. Solubility Test The solubility of SVS in water in combination with five different solubilizers was measured. Two polymers were used (i.e., PVP and Soluplus) and three surfactants (i.e., Pluronic, SLS, and Tween20). Four concentrations (m/V) (5%, 2%, 1%, and 0.5%) of each solubilizer were prepared and an excess of the drug was added to 10 mL of every solution. This was done in triplicate for every solubilizer concentration. The samples were put on a shaker at room temperature for three days, after which the solutions were filtered through a 0.21 µm pore size membrane filter and analyzed with high-performance liquid chromatography (HPLC) as described below. In addition, the solubility was determined in a water–ethanol (20:80) mixture with potential solubilizers (1% and 2.5% SLS, and 2.5% and 5% Soluplus) in order to investigate the influence of ethanol on the solubility, since the use of ethanol could reduce the amount of solubilizer. Twenty percent ethanol was applied since this is the highest concentration that can be safely used in the spray dryer device used in this study. 3.3. Preparation of the Materials The SVS:LYS molar ratio used is this study was 1:1. The selection of solubilizer and the amount needed to dissolve the drug were based on the solubility study. A predetermined amount of the solubilizer was dissolved in water and subsequently SVS and LYS were added to the solution. The solution was mixed until the components were dissolved. Spray drying was performed with a Büchi mini spray dryer (Model B-191, Büchi Labortechnik AG, Flawil, Switzerland). The following conditions were used: inlet temperature was 100 ˝ C, outlet temperature was adjusted to 45 ˝ C by adjusting the pump to 15% to give a flow rate of 3.9 mL/min, (see Table 2). Aspirator setting was always 100% and gas flow rate 600 L/h. The spray drying process of SVS-LYS from 5% SLS mixture with an inlet temperature of 110 ˝ C and 150 ˝ C was also investigated and in this case the pump setting and the flow rate varied between 18% and 20% and 4.3–4.7 mL/min, respectively, producing outlet temperatures between 50 and 65 ˝ C.

21543

Molecules 2015, 20, 21532–21548

SVS-LYS (molar ratio of 1:1) was also prepared by cryo-milling (CM) before physically mixing with a surfactant (the amount corresponding to what was present in the final product spray-dried from 5% SLS solution). The two milling chambers were filled with 500 mg SVS-LYS mixture and two 15 mm stainless steel balls. Milling was performed at 30 Hz in an oscillatory ball mill (Mixer Mill MM 400, Retch GmbH & Co., Haan, Germany). The total milling time was 60 min. Prior to milling and after every 10 min, the milling chambers were immersed in liquid nitrogen for 2 min to ensure cryogenic conditions. When the milling process was finished, the milling chambers were placed in a desiccator over silica. After they reached room temperature, the products were weighed and mixed with the correct amount of surfactant in a mortar. Physical mixtures (PMs) of the unprocessed starting materials were prepared by mixing with pestle in a mortar. 3.4. X-ray Powder Diffraction (XRPD) XRPD was performed using Bruker D8 Discover X-ray diffractometer (Karlsruhe, Germany) applying Cu Kα radiation (λ = 1.54 Å). Scanning of the samples was carried out between 5˝ and 35˝ 2θ with a step size of 0.011˝ and a scan speed of 0.1 s/step. An acceleration voltage of 40 kV and current of 40 mA were used. The data was collected by DIFFRAC.V3 program. 3.5. Elemental Analysis of Spray-Dried Particle Surfaces The spray-dried particles were analyzed with a Zeiss Sigma HD VP scanning electron microscope (SEM, Carl Zeiss Microscopy, Jena, Germany) equipped with two energy-dispersive X-ray (EDS) detectors (60 mm2 , Thermo Noran, Thermo Fisher Scientific, Waltham, MA, USA). The particles were fixed on an aluminum sample stub by using a carbon adhesive and subsequently scanned with an acceleration voltage of 6 keV under 5 Pa nitrogen atmosphere (except 10 Pa for SVS from 5% SLS and SVS-LY from 0.5% SLS). 3.6. Differential Scanning Calorimetry (DSC) The DSC analysis was performed using a Mettler Toledo DSC823e (Schwerzenbach, Switzerland) attached with a Julabo FT 900 Cooler. The DSC thermograms of the samples were obtained under a nitrogen gas flow of 50 mL/min. Aluminum pans (40 µL) were filled with sample powder (4–8 mg) and heated at a rate of 10 K/min from ´50 ˝ C to 220 ˝ C. Crystalline SLS was heated from 25 ˝ C to 220 ˝ C at 10 K/min and held for 5 min at the final temperature. Afterwards, the melt was immediately cooled to ´50 ˝ C, where it was held for 15 min. Thereafter, it was heated again from ´50 ˝ C to 250 ˝ C. STARe software was used to determine the glass transition temperatures (Tg , midpoint), melting temperatures (Tm , onset), and recrystallization temperatures (Trc , onset) of the samples. These were calculated as the mean of three independent measurements. 3.7. Fourier-Transform Infrared Spectroscopy (FTIR) Infrared spectroscopy was performed with a FTIR Thermo Nicolet Nexus 8700 FT-IR (Thermo Scientific, Madison, WI, USA) using an ATR (attenuated total reflectance) accessory. Thermo Scientific OMNIC software (version 6.0a, Thermo Scientific) was applied to collect the samples. The spectra (64 scans) were recorded in a wave number range of 550 to 4000 cm´1 with a resolution of 4 cm´1 . 3.8. Dissolution Testing Dissolution testing was performed using a Distek dissolution system 2100 C (North Brunswick, NJ, USA) with a paddle rotation speed of 50 rpm and temperature of 37 ˝ C. A powder amount equivalent to 20 mg of the drug was placed on the bottom of the dissolution chamber. The chamber was subsequently filled with 500 mL of the preheated dissolution medium. The dissolution medium was a phosphate buffer with pH 7.2 (USP). All formulations were measured in triplicate.

21544

Molecules 2015, 20, 21532–21548

The dissolution profiles were measured over a period of 24 h. At predetermined intervals (2, 4, 6, 10, 20, 30, 40, 60, 90, 120, 240, 360, 480, and 1440 min), 5-mL aliquots were taken from the chamber and immediately replaced with the same volume of phosphate buffer solution (pH 7.2). After filtering the samples through a 0.21 µm membrane filter, the filtrates were immediately diluted with acetonitrile (ACN, VWR International S.A.S., Fontenay-Sous Bois, France) to correspond to the mobile phase composition (see below). Further dilution, if necessary, was done by 70/30 ACN/H2 O. Samples were analyzed with high-performance liquid chromatography (HPLC) as described below. Single-factor ANOVA analysis was performed to investigate whether the dissolution profiles of two formulations were statistically different (95% confidence level). 3.9. High-Performance Liquid Chromatography (HPLC) Quantitative determination of the drug concentrations was performed using a Gilson HPLC analysis system with UV-VIS 151 detector (Gilson, Middleton, WI, USA), 234 auto-injector (Gilson, Villiers le Bel, France), 321 pump (Gilson, Villiers le Bel, France), system interface module (Gilson), and Unipoint TM LC system version 3.01 software (Gilson). A Phenomenex Gemini-NX 5 µm C18 110 Å (250 ˆ 4.6 mm) column and SecurityGuard precolumn (Phenomenex Inc., Torrance, CA, USA) were used. The drug concentrations were determined using a mobile phase consisting of 70% ACN, 30% ultrapure water, and 0.1% trifluoroacetic acid (TFA, Sigma-Aldrich, St. Louis, MO, USA). The mobile phase flow rate was 1.2 mL/min and the detection wavelength was 238 nm. Standard solutions with concentrations of 0.5, 1, 5, 10, 25, 50, and 100 µg/mL of the drug were prepared in 70/30 ACN/H2 O. 3.10. Stability Study The formulations selected for the stability study were stored at 4 ˝ C/0% relative humidity (RH), room temperature (25 ˝ C)/60%RH, and 40 ˝ C/0% RH. Phosphorous pentoxide was used to achieve the RH of 0% while the RH of 60% was obtained with a saturated NaBr solution. The samples were analyzed regularly with XRPD and FTIR to detect solid-state changes occurring in the samples. 4. Conclusions In this study, co-amorphous drug–amino acid mixtures were spray-dried form aqueous solutions for the first time by using a surface-active agent sodium lauryl sulfate (SLS) as a solubilizer for the poorly water-soluble drug simvastatin (SVS). SVS and lysine (LYS) were dissolved at a 1:1 molar ratio in 0.5% or 5% SLS water solutions, which were then spray-dried to obtain the formulations of SVS-LYS from 0.5% SLS and SVS-LYS from 5% SLS. In addition, a spray-dried formulation without LYS (SVS from 5% SLS) and a co-milled SVS-LYS reference, physically mixed with SLS, were prepared. Elemental analysis by energy-dispersive X-ray spectroscopy (EDS) revealed that SLS coated the particles that were formed upon spray drying. Solid-state analysis by X-ray powder diffraction (XRPD) and differential scanning calorimetry (DSC) revealed that in all the spray-dried formulations the remaining crystallinity originated from SLS only. However, the crystalline structure of SLS was modified in these mixtures, as revealed by changes in the shape of the XRPD diffraction peaks of SLS, melting behavior of SLS, and the SO3 asymmetric vibrational feature of SLS in the Fourier-transform infrared (FTIR) spectra of the samples. SVS from 5% SLS formulation showed the best dissolution properties in pH 7.2, as a “spring and parachute” effect was observed. The maximum dissolved amount of SVS, which was almost twice the maximum dissolved amount of the corresponding physical mixture, was reached after 90 min. Despite the presence of at least partially crystalline SLS in the mixtures, all the studied formulations were able to significantly extend the stability of amorphous SVS compared to what has previously been seen with co-amorphous formulations of SVS. The best stabilizing effect (at least 12 months in dry conditions) was found when SLS was spray-dried with SVS (and LYS) instead of physically mixing it with co-amorphous SVS-LYS. The stabilizing effect of SLS could be due to 21545

Molecules 2015, 20, 21532–21548

physical separation of the amorphous material, as SLS was found to coat the spray-dried particles. In addition, SLS might protect the amorphous content form ambient moisture. In conclusion, spray drying of SVS (and LYS) from aqueous surfactant solutions was able to produce formulations with improved physical stability for amorphous SVS. This study showed that particle coating with a surface active agent may be a promising way to stabilize amorphous materials. Acknowledgments: The authors would like to thank the Microscopy laboratory at SIB Labs, University of Eastern Finland for providing laboratory facilities and Jari Leskinen for microscopy operation. Author Contributions: R.L., K.L, H.G., and T.R. conceived and R.L. designed the experiments; G.C. performed the experiments; G.C. and R.L. analyzed the data; R.L. wrote the paper with contribution from K.L, H.G., and T.R. Conflicts of Interest: The authors declare no conflict of interest.

References 1.

2. 3. 4.

5. 6.

7.

8.

9. 10.

11. 12.

13.

14.

Williams, H.D.; Trevaskis, N.L.; Charman, S.A.; Shanker, R.M.; Charman, W.N.; Pouton, C.W.; Porter, C.J.H. Strategies to address low drug solubility in discovery and development. Pharmacol. Rev. 2013, 65, 315–499. [CrossRef] [PubMed] Brough, C.; Williams, R.O., III. Amorphous solid dispersions and nano-crystal technologies for poorly water-soluble drug delivery. Int. J. Pharm. 2013, 453, 157–166. [CrossRef] [PubMed] Vo, C.L.; Park, C.; Lee, B. Current trends and future perspectives of solid dispersions containing poorly water-soluble drugs. Eur. J. Pharm. Biopharm. 2013, 85, 799–813. [CrossRef] [PubMed] Zheng, E.; Jain, A.; Papoutsakis, D.; Dannenfelser, R.M.; Panicucci, R.; Garad, S. Selection of oral bioavailability enhancing formulations during drug discovery. Drug. Dev. Ind. Pharm. 2012, 38, 235–247. [CrossRef] [PubMed] Janssens, S.; van den Mooter, G. Review: physical chemistry of solid dispersions. J. Pharm. Pharmacol. 2009, 61, 1571–1586. [CrossRef] [PubMed] Srinarong, P.; de Waard, H.; Frijlink, H.W.; Hinrichs, W.L.J. Improved dissolution behavior of lipophilic drugs by solid dispersions: The production process as starting point for formulation considerations. Expert Opin. Drug Deliv. 2011, 8, 1121–1140. [CrossRef] [PubMed] Laitinen, R.; Löbmann, K.; Grohganz, H.; Strachan, C.; Rades, T. Amino acids as co-amorphous excipients for simvastatin and glibenclamide: Physical properties and stability. Mol. Pharm. 2014, 11, 2381–2389. [CrossRef] [PubMed] Löbmann, K.; Strachan, C.; Grohganz, H.; Rades, T.; Korhonen, O.; Laitinen, R. Co-amorphous simvastatin and glipizide combinations show improved physical stability without evidence of intermolecular interactions. Eur. J. Pharm. Biopharm. 2012, 81, 159–169. [CrossRef] [PubMed] Laitinen, R.; Löbmann, K.; Strachan, C.J.; Grohganz, H.; Rades, T. Emerging trends in the stabilization of amorphous drugs. Int. J. Pharm. 2013, 30, 65–79. [CrossRef] [PubMed] Löbmann, K.; Grohganz, H.; Laitinen, R.; Strachan, C.; Rades, T. Amino acids as co-amorphous stabilizers for poorly water soluble drugs-Part 1: Preparation, stability and dissolution enhancement. Eur. J. Pharm. Biopharm. 2013, 85, 873–881. Gao, Y.; Liao, J.; Qi, X.; Zhang, J. Coamorphous repaglinide-saccharin with enhanced dissolution. Int. J. Pharm. 2013, 450, 290–295. [CrossRef] [PubMed] Hu, Y.; Gniado, K.; Erxleben, A.; McArdle, P. Mechanochemical reaction of sulfathiazole with carboxylic acids: Formation of a cocrystal, a salt, and coamorphous colids. Cryst. Growth. Des. 2014, 14, 803–813. [CrossRef] Dengale, S.J.; Ranjan, O.P.; Hussen, S.S.; Krishna, B.S.M.; Musmade, P.B.; Shenoy, G.G.; Bhat, K. Preparation and characterization of co-amorphous ritonavir–indomethacin systems by solvent evaporation technique: Improved dissolution behavior and physical stability without evidence of intermolecular interactions. Eur. J. Pharm. Sci. 2014, 62, 57–64. [CrossRef] [PubMed] Shayanfar, A.; Jouyban, A. Drug-drug coamorphous systems: Characterization and physicochemical properties of coamorphous atorvastatin with carvedilol and glibenclamide. J. Pharm. Innov. 2013, 8, 218–228. [CrossRef]

21546

Molecules 2015, 20, 21532–21548

15.

16.

17.

18.

19. 20.

21. 22. 23.

24. 25. 26. 27. 28. 29. 30. 31. 32.

33.

34.

35. 36.

Martínez, L.M.; Videa, M.; López-Silva, G.A.; de los Reyes, C.A.; Cruz-Angeles, J.; González, N. Stabilization of amorphous paracetamol based systems using traditional and novel strategies. Int. J. Pharm. 2014, 477, 294–305. [CrossRef] [PubMed] Jensen, K.T.; Blaabjerg, L.I.; Lenz, E.; Bohr, A.; Grohganz, H.; Kleinebudde, P.; Rades, T.; Löbmann, K. Preparation and characterization of spray-dried co-amorphous drug–amino acid salts. J. Pharm. Pharmacol. 2015. [CrossRef] [PubMed] Démuth, B.; Nagy, Z.K.; Balogh, A.; Vigh, T.; Marosi, G.; Verreck, G.; van Assche, I.; Brewster, M.E. Downstream processing of polymer-based amorphous solid dispersions to generate tablet formulations. Int. J. Pharm. 2015, 486, 268–286. [CrossRef] [PubMed] Koltzenburg, S. Formulation of problem drugs –and they are all problem drugs. In Solubility Enhancement with BASF Pharma Polymers. Solubilizer Compendium; Reintjes, T., Ed.; BASF SE: Lampertheim, Germany, 2011; p. 22. Mukerjee, P.; Mysels, K.J. Critical micelle concentrations of aqueous surfactant systems. Nat. Stand. Ref. Data Ser. Nat. Bur. Stand. 1970. [CrossRef] Patist, A.; Bhagwat, S.S.; Penfield, K.W.; Aikens, P.; Shah, D.O. On the measurement of critical micelle concentrations of pure and technical-grade nonionic surfactants. J. Surfactants Deterg. 2000, 3, 53–58. [CrossRef] Available online: http://www4.mpbio.com/ecom/docs/proddata.nsf/03d5991c3a7e7f0d8525792e00243 e31/b8dd4c6faf0859048525663c0070508e (accessed on 20 November 2015). Cadwallader, D.E.; Devendra, K.M. Effect of macromolecules on aqueous solubility of cholesterol and hormone drugs. J. Pharm. Sci. 1981, 70, 442–446. [CrossRef] [PubMed] Garekani, H.A.; Sadeghi, F.; Ghazi, A. Increasing the aqueous solubility of acetaminophen in the presence of polyvinylpyrrolidone and investigation of the mechanisms involved. Drug Dev. Ind. Pharm. 2003, 29, 173–179. [CrossRef] [PubMed] Ambike, A.A.; Mahadik, K.R.; Paradkar, A. Spray-dried amorphous solid dispersions of simvastatin, a low Tg drug: In vitro and in vivo evaluations. Pharm. Res. 2005, 22, 990–998. [CrossRef] [PubMed] Zhang, Y.; Liu, Y.; Luo, Y.; Yao, Q.; Zhong, Y.; Tian, B.; Tang, X. Extruded soluplus/SIM as an oral delivery system: Characterization, interactions, in vitro and in vivo evaluations. Drug Deliv. 2015. [CrossRef] Singla, N.; Gupta, G.D.; Kohli, K.; Singla, A.K. A discriminatory and biorelevant dissolution test method for simvastatin drug products. Dissol. Technol. 2009, 16, 11–13. [CrossRef] Takeuchi, H.; Nagira, S.; Yamamoto, H.; Kawashima, Y. Solid dispersion particles of tolbutamide with fine silica particles by the spray-drying method. Powder Technol. 2004, 195, 141–187. [CrossRef] Yu, D.G.; Zhu, L.M.; Branford-White, C.J.; Yang, J.H.; Wang, X.; Li, Y.; Qian, W. Solid dispersions in the form of electrospun core-sheath nanofibers. Int. J. Nanomed. 2012, 6, 3271–3280. [CrossRef] [PubMed] Roberts, M.; Ehtezazi, T.; Compernolle, A.; Amin, K. The effect of spray drying on the compaction properties of hypromellose acetate succinate. Drug Dev. Ind. Pharm. 2011, 37, 268–273. [CrossRef] [PubMed] Millqvist-Fureby, A.; Malmsten, M.; Bergenståhl, B. Spray-drying of trypsin—Surface characterisation and activity preservation. Int. J. Pham. 1999, 188, 243–253. [CrossRef] Sinko, P.J. Martin’s Physical Pharmacy and Pharmaceutical Sciences, 6th ed.; Lippincott Williams & Wilkins: Baltimore, MD, USA, 2011; p. 369. Mattern, M.; Winter, G.; Kohnert, U.; Lee, G. Formulation of proteins in vacuum-dried glasses. II. Process and storage stability in sugar-free amino acid systems. Pharm. Dev. Technol. 1999, 4, 199–208. [CrossRef] [PubMed] Dave, R.H.; Patel, A.D.; Donahue, E.; Patel, H.H. To evaluate the effect of addition of an anionic surfactant on solid dispersion using model drug indomethacin. Drug Dev. Ind. Pharm. 2012, 38, 930–939. [CrossRef] [PubMed] Nasr, H.E.; Gaffar, M.S.; Abdel-Kareem, O.; Abd El-Aziz, F. Morphological, rheological and ultrasonic characterizations of ECO-friendly microemulsion lattices based on acrylate monomers. J. Am. Sci. 2010, 6, 897–910. Da Costa, M.A.; Seiceira, R.C.; Rodrigues, C.R.; Hoffmeister, C.R.D.; Cabral, L.M.; Rocha, H.V.A. Efavirenz dissolution enhancement I: Co-micronization. Pharmaceutics 2013, 5, 1–22. [CrossRef] [PubMed] Bhise, S.B.; Mookkan, S. Formulation and evaluation of novel FDCs of antitubercular drugs. J. Pharm. Res. 2009, 2, 437–444.

21547

Molecules 2015, 20, 21532–21548

37.

38.

39.

40. 41. 42. 43. 44. 45.

46. 47.

48.

49.

Kaminska, E.; Adrjanowicz, K.; Kaminski, K.; Wlodarczyk, P.; Hawelek, L.; Kolodziejczyk, K.; Tarmacka, M.; Zakowiecki, D.; Kacmarcyk-Sedlak, I.; Pilch, J.; et al. A new way of stabilization of furosemide upon cryogenic ginding by using acylated saccharides matrices. The role of hydrogen bonds in decomposition mechanism. Mol. Pharmacol. 2013, 10, 1824–1835. [CrossRef] [PubMed] Grzybowska, K.; Paluch, M.; Wlodarczyk, P.; Kaminski, K.; Hawelek, L. Enhancement of amorphous celecoxib stability by mixing it with octaacetylmaltose: the molecular dynamics study. Mol. Pharm. 2012, 9, 894–904. [CrossRef] [PubMed] Marsac, P.J.; Li, T.; Taylor, L.S. Estimation of drug-polymer miscibility and solubility in amorphous solid dispersions using experimentally determined interaction parameters. Pharm. Res. 2009, 26, 139–151. [CrossRef] [PubMed] Marsac, P.J.; Konno, H.; Taylor, L.S. A comparison of the physical stability of amorphous felodipine and nifedipine systems. Pharm. Res. 2006, 23, 2306–2316. [CrossRef] [PubMed] Mosquera-Giraldo, L.I.; Trasi, N.S.; Taylor, L.S. Impact of surfactants on the crystal growth of amorphous celecoxib. Int. J. Pharm. 2014, 461, 251–257. [CrossRef] [PubMed] Prosser, A.J.; Franses, E.I. Infrared reflection absorption spectroscopy (IRRAS) of aqueous nonsurfactant salts, ionic surfactants, and mixed ionic surfactants. Langmuir 2002, 18, 9234–9242. [CrossRef] Viana, R.B.; da Silva, A.B.F.; Pimentel, A.S. Infrared spectroscopy of anionic, cationic, and zwitterionic surfactants. Adv. Phys. Chem. 2012. [CrossRef] Kawai, T.; Umemura, J.; Takenaka, T. Fourier transform infrared study on the phase transitions of a sodium dodecyl sulfate-water system. Bull. Inst. Chem. Res. Kyoto Univ. 1983, 61, 314–323. Heikkinen, A.T.; DeClerck, L.; Löbmann, K.; Grohganz, H.; Rades, T.; Laitinen, R. Dissolution properties of co-amorphous drug-amino acid formulations in buffer and biorelevant media. Die Pharm. 2015, 70, 452–457. Augustijns, P.; Brewster, M.E. Supersaturating drug delivery systems: Fast is not necessarily good enough. J. Pharm. Sci. 2012, 101, 7–9. [CrossRef] [PubMed] Chen, J.; Ormes, J.D.; Higgins, J.D.; Taylor, L.S. Impact of surfactants on the crystallization of aqueous suspensions of celecoxib amorphous solid dispersion spray dried particles. Mol. Pharm. 2015, 12, 533–541. [CrossRef] [PubMed] Priemel, P.A.; Laitinen, R.; Barthold, S.; Grohganz, H.; Lehto, V.P.; Rades, T.; Strachan, C.J. Inhibition of surface crystallisation of amorphous indomethacin particles in physical drug-polymer mixtures. Int. J. Pharm. 2013, 456, 301–306. [CrossRef] [PubMed] Collier, J.W.; Shah, R.B.; Gupta, A.; Sayeed, V.; Habib, M.J.; Khan, M.A. Influence of formulation and processing factors on stability of levothyroxine sodium pentahydrate. AAPS PharmSciTech 2010, 11, 818–825. [CrossRef] [PubMed]

Sample Availability: Samples of the compounds are not available from the authors. © 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).

21548